Mitochondrial alterations in embryos exposed to B-hydroxybutyrate in whole embryo culture.код для вставкиСкачать
THE ANATOMICAL RECORD 213:94-lOl(1985) Mitochondria1Alterations in Embryos Exposed to B-Hydroxybutyrate in Whole Embryo Culture W.E. HORTON, JR. AND T.W. SADLER Department of Anatomy, School of Medicine, University of North Carolina, Chapel Hill, NC 27514 ABSTRACT The ketone body B-hydroxybutyrate (B-OHB) produces malformations and ultrastructural alterations in mitochondria of mouse embryos exposed for 24 hours to the compound in whole embryo culture. The present study was conducted to establish the time-course of the mitochondrial changes to determine whether the changes are reversible, and to relate these changes to the malformations produced by the compound. Since mitochondria also play a key role in the metabolism of ketone bodies, the capacity of the early somite embryo to metabolize B-OHB was investigated in a n effort to link the morphological alterations in the mitochondria to a biochemical process. Early somite embryos were cultured 4, 8, or 24 hours in the presence of 32 mM DL-B-OHB and then cultured for a n additional 24 hours in control serum. Finally, embryonic tissue during the teratogenic period was assessed for its capability to oxidize B-OHB using D-(3-14C)-B-OHB.The treated embryos showed progressive alterations in the mitochondria, beginning at 4 hours with a loss of matrix density and culminating a t 24 hours with high-amplitude swelling, complete loss of matrix density, and disappearance of cristae. These alterations were reversible following removal of the embryos after 24 hours of exposure to B-OHB and culturing for a n additional 24 hours in control serum. Metabolism studies demonstrated that the early somite embryo possesses a limited capacity to oxidatively metabolize B-OHB. The biochemical implications of these findings are discussed with respect to the possible role of ketone bodies in the mechanism of diabetes-induced congenital malformations. Pregnancy in the human diabetic is fraught with many complications including a n increased incidence of congenital malformations in the offspring (Pederson, 1980). Although many factors may be responsible for these defects, recent evidence suggests that elevated levels of ketone bodies (specifically B-hydroxybutyrate [B-OHB]) may play a role (McLendon and Bottomy, 1960; Drury, 1966).Two separate studies using whole embryo culture techniques have shown that increased concentrations of DL-B-OHB are teratogenic to both mouse (Horton and Sadler, 1983) and rat (Lewis et al., 1983) embryos. The embryonic response was dose and age related and was associated with growth retardation. Histological and ultrastructural analysis of malformed embryos revealed a single alteration in cellular morphology manifested by high-amplitude swelling of mitochondria (Horton and Sadler, 1983). The relationship between this alteration and the congenital malformations is not known, although fluctuations in mitochondrial morphology have been attributed to both normal (Hackenbrock, 1966, 1968) and pathological conditions, notably cell necrosis (Laiho and Trump, 1975). The mechanism by which B-OHB produces mitochondrial swelling in the intact embryo as well as the progression and time-course of the mitochondrial changes are not known. Furthermore, the relationship between the effects on mitochondria and cell viability has not 0 1985 ALAN R. LISS, INC been established. Also, the ability of the mitochondria (and whole embryos) to recover from B-OHB treatment has not been determined. Finally, the biochemical capabilities of the early- somite embryo to oxidatively metabolize B-OHB, a process that is initiated in the mitochondria, during the sensitive teratogenic period has not been assessed. Therefore, the following study was undertaken. METHODS Embryo Culture Random-bred ICR mouse embryos were obtained on the ninth gestational day (plug day = day 1)and prepared for whole embryo culture as described previously (New, 1978; Sadler, 1979). Individual embryos (three to five somites in age) were placed in 30-ml flasks containing 3 ml of whole, immediately centrifuged (Steele, 1972) rat serum. Each flask was gassed for approximately 30 seconds with 5% 0 2 , 5% C02, and 90% N2 at 0 and approximately 10 hours and placed on a rotator wheel (30 rpm) in a 38°C incubator. For embryos maintained longer than 24 hours, cultures were transferred to fresh serum a t 24 hours and gassed with a 20% 0 2 , 5 % C02, and 75% N2 mixture a t 24 and 34 hours. Received July 16, 1984; accepted November 9, 1984. 3-OH-BUTYRATE-INDUCED CHANGES IN EMBRYOS Upon termination of culture, embryos were examined for gross abnormalities and analyzed for total protein content using the Lowry method (Lowry et al., 1951) or prepared for light and electron microscopic observation by fixing in modified Karnovsky’s (2% glutaraldehyde, 2% paraformaldehyde), postfixing in 1% osmium tetroxide, dehydrating in alcohol and propylene oxide, and embedding in araldite. Embedded embryos were sectioned at 1 pm with a n LKB ultramicrotome V and stained with Toluidine blue for histological examination. In addition, thin sections were cut through cephalic regions, stained with uranyl acetate (1%in ethanol)-lead citrate, and examined with a JEOL 100-CX electron microscope. Exposure to B-OHB A stock solution of DL-B-OHB sodium salt (Sigma Chemical Co., St. Louis) was prepared in distilled water and added to rat serum to achieve a final concentration of 4 mg/ml (32 mM). This dose has previously been shown to produce malformations and mitochondrial swelling in a high percentage of cultured embryos (Horton and Sadler, 1983). Embryos were cultured in this serum or in serum containing distilled water only for 4, 8, or 24 hours, then terminated and examined as described above. In addition, one group of embryos was cultured for 24 hours in the presence of 32 mM B-OHB, then rinsed three times in Tyrode’s solution and transferred to fresh control serum for a n additional 24 hours of culture. A minimum of 14 embryos (seven treated and seven controls) were cultured at each time period and were examined with electron microscopy. Protein data for the 48-hour time period were compared using a oneway analysis of variance. B-OHB Metabolism Ketone body utilization was examined on day 9 of gestation by determining the production of I4CO2 from D-(3-14C)B-OHB in vitro (Shambaugh et al., 1977a). Whole conceptus (embryo and yolk sac) were pooled in ice-cold PBS, weighed, and placed in flasks with 1.5 ml whole, immediately centrifuged rat serum containing 4mM DL-B-OHB and 0.1 uCi/ml D-(3-14C)-B-OHB (Amersham, specific activity = 55 mCi/mM). The tissue was gassed with 5% 0 2 , 5% COZ 90% Nz, prior to a 4hour incubation a t 38°C. Reactions were then stopped by placing flasks on ice and adding 0.3 ml hyamine hydroxide to the center well and 0.3 ml 6 N Hz SO4 to the serum and continuing to rotate the flasks for 1hour at 38°C to evolve the CO2. The hyamine hydroxide was then transferred from the center well to scintillation vials containing 10 ml of Aquasol 11. The activity of the hyamine fraction was converted to pmoles of COB evolved from B-OHB by determining on a percent basis the amount of labelled B-OHB utilized and assuming and equal rate of utilization for the unlabelled compound. For comparison, maternal liver was assayed for B-OHB utilization as described above. Each study (maternal liver and day 9 conceptus) was repeated three times. 95 development (Figs. 1, 2). Cranial neural folds at these time periods were in the initial stages of elevation and consisted of a pseudostratified epithelium with three to four nuclear layers. The luminal surface was uniform in appearance with adjacent cells joined by terminal bars, while a basement membrane was present on the abluminal side. Numerous mitotic figures were observed along the luminal surface and were characterized by spherical nuclei containing condensed chromatin. Interphase nuclei were variable in shape and contained one to three prominent nucleoli. An occasional pyknotic cell was present, but there was no evidence of extensive cell necrosis or other cellular abnormalities. Beneath the basement membrane of the neuroepithelium was a region of embryonic mesenchyme cells. Each mesenchymal cell was composed of a nuclear region (some in mitosis) and surrounding cytoplasmic projections which often contacted other cells. The cytoplasm of these cells was uniform in appearance and no evidence of cell necrosis was observed. Ultrastructurally, neuroepithelial cells were similar in appearance to mesenchymal and gut endoderm tissue, all resembling undifferentiated embryonic cell types (Figs. 3,4). The cytosome of these cells was characterized by a polysomal arrangement of the many ribosomes present. Also observed were a few cisternae of rough endoplasmic reticulum and a n occasional Golgi complex. Numerous mitochondria were distributed in a random fashion throughout the cells. These displayed distinct inner and outer membranes, relatively few and indistinct cristae, and a homogenous dense matrix. Typical cross sections of these organelles ranged from 0.5 to 1.0 Dm in diameter. Four-Hour-TreatedEmbryos After 4 hours of exposure to 32 mM DL-B-OHB, embryos displayed no gross abnormalities. Light microscopy revealed a histological picture similar to controls in most respects (Fig. 5). The neuroepithelium was intact with a smooth luminal surface, and mesenchymal tissue revealed no abnormal morphology and no cell necrosis was observed. However, cytoplasmic vacuoles were consistently observed in cells of the neuroepithelium, mesenchyme, and endoderm. These vacuoles were spherical in shape and ranged in size from 0.5 to 1.5 pm. They were numerous in some cells and completely absent in others. From previous work (Horton and Sadler, 1983) it was known that the cytoplasmic vacuoles represented mitochondria in various degrees of swelling. Therefore, electron microscopic examination of 4-hour-treated embryos was performed to characterize the altered mitochondrial morphology. In general, most of these organelles appeared similar to controls with a dense matrix and relatively indistinct cristae. At the other extreme, mitochondria were observed that showed definite swelling and these organelles displayed a pale homogenous matrix. In some cases cristae were present (more distinct against the pale matrix than in controls) while other examples displayed few or no cristae. A third morpholRESULTS ogy was observed at this time period, in which the miFour- and Eight-Hour Control Embryos tochondria appeared in the initial stages of swelling Control embrvos examined with light microscopy a t 4 (Fig. 6). These organelles displayed a dense matrix interand 8 hours oiculture displayed a-similar patiern of sp&sed with isdated areas of pale matrix, although 96 W.E. HORTON, JK.AND T.W. SADLER Fig. 1. Cross section through an early somite embryo cultured 4 hours in control serum. NF, cranial neural fold; H, heart; YS, visceral yolk sac. x23. Fig. 3. Portion of a neuroepithelial cell from an embryo cultured 4 hours in control serum. Numerous mitochondria (M) cut in cross section and longitudinally are dispersed throughout the cell. x 19,800. Fig. 2. Neuroepithelium from embryo cultured 4 hours in control serum (a similar morphology is present after 8 hours in culture). The neuroepithelium WE) is psuedostratified with three to four nuclear layers. Mitotic figures are observed (arrow) and no abnormal cytoplasmic structures are evident. A, amnion; M, mesenchyme. ~ 9 0 0 . Fig. 4. High-magnification view of typical mitochondria from a control embryo. Note the dense matrix and the relative indistinct appearance of the cristae (C). Also, observe the polysomal arrangement of the ribosomes (P). ~40,700. Fig. 5. Neuroepithelium (NE) from an embryo cultured 4 hours in the presence of 32 mM DL-B-OHB. Note the presence of numerous, small, cytoplasmic vacuoles (arrows). A, amnion; M, mesenchyme. x660. Fig. 8 . Electron micrograph of a neuroepithelial cell from the embryo described in Figure 7. Note the swollen mitochondria (M).C,cristae; P, polysomes. x 30,500. Fig. 6. Electron micrograph of a neuroepithelial cell from the embryo described in Figure 5. Note the electron-lucent portions of the mitochondrial matrix (arrows). P, polysomes. x 30,500. Fig. 9. Neuroepithelium (NE) from a n embryo cultured 24 hours in the presence of 32 mM DL-B-OHB. Cytoplasmic vacuoles are widespread and many are larger than those observed at earlier time periods (arrows). x660. Fig. 7. Neuroepithelium (NE) from an embryo cultured 8 hours in the presence of 32 mM DL-B-OHB.Cytoplasmic vacuoles (arrows) are prevalent. “Blebbing” of the luminal surface of the neuroepithelium has occurred (arrowheads, compare to Fig. 5). X660. Fig. 10. Electron micrograph of a neuroepithelial cell from the embryo described in Figure 9. A single, highly swollen mitochondrion is present (M) with a pale matrix and reduced cristae (C). P, polysomes. ~30,500. 98 W.E. HORTON, JR.AND T.W. SADLER there was no apparent increase in the overall size of these structures. No other ultrastructural abnormalities were observed at this time period. Eight-Hour-TreatedEmbryos Electron microscopy was performed on 48-hour control and treated tissue to assess the degree of recovery of mitochondria after exposure to DL-B-OHB. Interestingly, both control and treated tissue displayed similar profiles of mitochondrial morphology (Figs. 13, 141, although the appearance was distinct from that observed in control embryos a t earlier time periods. The control and “recovered” mitochondria each displayed a prominent inner and outer membrane. The inner membrane continued into the matrix region as cristae, which appeared more distinct than a t earlier time periods. Matrix morphology was also different from control tissue cultured for 4 and 8 hours. For example, earlier the matrix was electron-dense, whereas mitochondria from embryos (treated and controls) cultured for 48 hours exhibited a pale matrix consisting of a fine, homogeneous material. This pattern was especially evident in mitochondria that had been cut in a longitudinal profile. As with the earlier time period, no gross abnormalities were observed in embryos exposed 8 hours to DL-BOHB. At the light microscopic level, the most striking finding was a n increase in the size and number of cytoplasmic vacuoles. These structures were now present in virtually every cell and in some cases reached 2 pm in diameter (Fig. 7). Also, a change in the appearance of the luminal surface of the neuroepithelium was observed. Instead of the smooth configuration observed in 4-hour-treated embryos and all controls, the luminal surface was now irregular and characterized by “blebbing” of the surface cytoplasm. Electron microscopy revealed a progression in the mitochondrial response to B-OHB compared to the 4-hour Metabolism of B-OH6 time period. The majority of the mitochondria now disThe ability of developing tissue to oxidatively metabplayed high-amplitude swelling with loss of matrix density and few identifiable cristae (Fig. 8). In spite of this olize B-OHB to GO2 on day 9 of gestation was detersevere alteration in mitochondrial morphology, no other mined. At this stage (three to five somites), the mouse conceptus produced 0.84 & 0.26 pM COz/g wet weight/ ultrastructural changes were observed. minute from B-OHB. This value was slightly higher Twenty Four-Hour-Treated Embryos than the level determined for maternal liver (0.64 k Embryos cultured 24 hours in 32 mM DL-B-OHB dis- 0.34 pM C02/g wet weightlminute) which is known to played gross defects, as described previously (Horton possess only a slight capacity to metabolize the comand Sadler, 1983). These included cephalic and caudal pound (Mahler, 1953) and, therefore, served as a referneural tube closure defects, abnormal rotation, and ence for comparison with embryonic metabolic rates. growth reduction. Light histology revealed only minimal cell necrosis, and the presence of widespread cytoDISCUSSION plasmic vacuoles throughout the tissue (Fig. 9). The present study represents a n extension of previous Ultrastructurally these swollen mitochondria appeared more severely affected than at the 8-hour time period work dealing with interactions between the ketone body, (Fig. 10). The majority of the organelles were widely B-OHB, and the developing embryo in vitro. The results dilated (up to 3 pm in diameter) with a n extremely pale show that high concentrations of DL-B-OHB (32 mM) matrix. For the most part, normal cristae were not ap- induce progressive mitochondrial changes in mouse emparent. As with all other time periods, the polysomal bryos growing in whole embryo culture. Specifically, arrangement of ribosomes was intact and no other ultra- mitochondria undergo high-amplitude swelling, which begins by 4 hours of exposure to B-OHB and becomes structural abnormalities were noted. increasingly pronounced during the 24-hour culture peForty Eight-Hour Recovery Studies riod. This swelling occurs in the absence of ultrastrucControl embryos cultured for 48 hours were similar in tural alterations in other cellular organelles and is appearance to comparably staged in vivo embryos (Fig. reversible if embryos are cultured for a n additional 24 11).These embryos possessed 29-31 somites, had ad- hours in control serum. The specific mechanism by which B-OHB produces vanced forelimb development, and displayed a completely closed neural tube. Furthermore, the cranial mitochondrial swelling in the intact embryo is not region showed brain vesicle development resulting in known. Mitochondria have been shown to undergo morsubdivision of the neural tube into forebrain, midbrain, phological alterations in response to a variety of physiological, pathological, and physical conditions. For and hindbrain. Embryos cultured for 24 hours in the presence of 32 example, isolated mitochondria behave as osmometers, mM DL-B-OHB, followed by 24 hours in control serum shrinking and swelling in response to either a hyperdisplayed abnormal development compared to controls tonic or hypotonic medium, respectively (Lehninger, (Fig. 12). The majority of the treated embryos (92%)had 1962). In this regard, the addition of 32 mM levels of Bcompleted cranial neural tube closure and averaged 28- OHB raises the serum osmolarity approximately 40 30 somites. However, 50%(13/26)of the embryos showed mOsmAiter, i.e. from 300 to 340. However, this increase lack of closure of the posterior neuropore. Also, treated is not significant considering that embryos cultured 24 embryos consistently failed to undergo normal brain hours in the presence of 64-mM levels of L-glucose (which vesicle formation. These embryos displayed a rounded produces serum osmolarity levels in excess of 340 m O s d head with lack of expansion of the prosencephalon. Fi- liter) grow normally and show no cytoplasmic vacuoles nally, treated embryos were growth reduced, averaging (unpublished results). These results confirm earlier work 193 f 32 pg (N = 13)protein compared to 312 k 69 k g which demonstrated that high concentrations of L-glucose which produced osmolarities in excess of 390 mOsm/ (N = 12)protein for controls (P < .01). 3-OH-FUTYRATE-INDUCEDCHANGES IN EMBRYOS 99 Fig. 11. An embryo cultured 48 hours in control serum. H, heart; L, forelimb; 0, otic vesicle; P, prosencephalon; M, mesencephalon; R, rhombencephalon. ~ 2 6 . Fig. 13. Typical mitochondria from an embryo cultured 48 hours in control serum. Compare to Figure 4 and note paler matrix and more distinct cristae (C). P, polysomes. x 39,000. Fig. 12. An embryo cultured 24 hours in the presence of 32 mM DLB-OHB and then 24 hours in control serum. The embryo shows an overall growth reduction and retarded brain development, especially noticeable by the lack of prosencephalic expansion. X26. Fig. 14. Typical mitochondria from an embryo exposed to 32 rnM BOHB and then transferred to control serum for an additional 24 hours. The mitochondria1 morphology appears similar to that observed in controls with a pale matrix and distinct cristae (C). P, polysomes. x 39.000. 100 W.E. HORTON, JR.AND T.W. SADLER liter were not teratogenic to cultured rat embryos (Cockroft and Coppola, 1977). It is also possible that the passage of DL-B-OHB into the embryo and, ultimately, the mitochondria results in fluid movement and mitochondrial swelling. Yet, the time course of the swelling as revealed by electron microscopy (occurring somewhere between 4 and 8 hours) would seem too slow for what should be a rapid osmotic event. Also, fluid movement into the cell would be expected to cause more general morphological alterations, such as dilution of the cytosol, dilation of endoplasmic reticulum, or, possibly, nuclear changes. However, the ultrastructural studies demonstrated no differences between control and treated tissue except for the mitochondria] alterations. In addition to osmotic responses, mitochondrial morphology is also influenced by fixation. For example, low concentrations of paraformaldehyde (< 1%)have been shown to induce mitochondrial swelling in liver tissue (Romert and Matthiessen, 1981). However, it is unlikely that the results observed in this study are due to nonspecific effects of fixation. The concentration of paraformaldehyde used in this study was well above the level that produces swelling and, in fact, no abnormal mitochondria were noted in control tissues which were processed simultaneously with treated specimens. Also, embryos exposed to B-OHB and then fixed in 4% osmium tetroxide only displayed the cytoplasmic vacuoles indicative of mitochondrial swelling (unpublished results). Several investigators have observed high-amplitude mitochondrial swelling in association with cell injury and cell death (Laiho and Trump, 1975).Although some cell necrosis is visible after 24-hour exposure to B-OHB, the mitochondrial change does not seem to be linked to cell death. Mitochondrial swelling occurs in the majority of cells of the embryo (dying and non-dying), whereas only minimal cell necrosis is observed. Also the mitochondrial change occurs in the absence of other ultrastructural markers for cell death such as dispersal of polysomes (Sadler and Cardell, 1977). Finally, the mitochondrial swelling is fully reversible if B-OHB is removed from the culture medium. Mitochondrial changes induced by B-OHB might also have a biochemical basis. The enzyme pathway for the oxidative metabolism of ketone bodies is intramitochondrial, involving the conversion of B-OHB to acetoacetate by the stereospecific enzyme B-OHB dehydrogenase (Robinson and Williamson, 1980). It has been postulated that the shunting of B-OHB through this metabolic pathway decreases intramitochondrial NAD+/NADH ratios and may lead to alterations in mitochondrial function (Goldstein et al., 1982). The present study has demonstrated that the early somite embryo can metabolize B-OHB to C 0 2 a t a rate slightly greater than that obtained for maternal liver which minimally utilizes ketone bodies (Mahler, 1953). Thus, it is possible that BOHB is undergoing initial stages of metabolism leading to alterations in redox potential of the mitochondria. In addition to their association with mitochondria, BOHB andor its metabolites are known to interact with several other biochemical pathways, including inhibition of (1)uptake and utilization of glucose by adult and fetal tissues (Robinson and Williamson, 1980; Shambaugh et al., 197713); (2) the oxidative metabolism of alpha-ketoisocaproic acid (Shambaugh and Koehler, 1983);and (3)pyrimidine synthesis in the fetal rat brain (Bhasin and Shambaugh, 1982). This last interaction has been implicated as a potential mechanism by which ketone bodies might inhibit cell proliferation and lead to the lowered brain weight in offspring of women experiencing ketonemic states late in pregnancy (Bhasin and Shambaugh, 1982). In the present study, embryos cultured for 24 hours in B-OHB and 24 hours in control serum displayed abnormal brain development with lack of brain vesicle formation, and were growth reduced. Also, mitotic indices performed on embryos treated 24 hours with 32 mM DLBOHB revealed a 30% decrease in mitotic rate (unpublished results). Thus, it is possible that the effects of BOHB on the developing embryo may be mediated via a n effect on cell proliferation. Studies are currently underway in our laboratory to further assess the effects of B-OHB on pyrimidine metabolism and to determine the effects of acetoacetate on embryogenesis and mitochondrial morphology. It is hoped that these types of studies will lead to a better understanding of possible biochemical alterations produced by ketone bodies in the developing embryo and, thus, clarify potential mechanisms of diabetes-induced congenital malformations. ACKNOWLEDGMENTS This work was supported in part by the Kroc Foundation, the Ryan Foundation, and NIH grant #HD 17381. The authors greatly appreciate the technical assistance of Julie Yonker and Sheryl Tulis, and the typing assistance of Robin Wynn. LITERATURE CITED Bhasin, S., and G.E. Shambaugh 111(1982)Fetal fuels V. Ketone bodies inhibit pyrimidine biosynthesis in fetal rat brain. Am. J. Physiol., 243:E234-E239. Cockroft, D.L. and P.T. Coppola (1977) Teratogenic effects of C X C ~ S S glucose on head-fold rat embryos in culture. Teratology, 16:141146. Drury, M.I. (1966)Pregnancy in the diabetic. Diabetes, 15:830-835. Goldstein, L., R.J. Solomon, D. Pearlman, P.M. McLaughlin, and M.A. Taylor (1982) Ketone body effects on glutamine metabolism in isolated kidneys and mitochondria. Am. J. Physiol., 243:F181F187. Hackenbrock, C.R. (1966)Ultrastructural basis for metabolically linked mechanical activity in mitochondria I. Reversible ultrastructural changes with change in metabolic steady state in isolated liver mitochondria. J. Cell Biol., 30:269-297 Hackenbrock, C.R. (1968)Ultrastructural basis for metabolically linked mechanical activity in mitochondria 11. Electron transport-linked ultrastructural transformations in mitochondria. J. Cell Biol., 37:345-362. Horton, W.E., and T.W. Sadler (1983) Effects of maternal diabetes on early embryogenesis: Alterations in morphogenesis produced by the ketone body, B-hydroxybutyrate. Diabetes, 32610-616. Laiho, K.U., and B.F. Trump (1975)Studies on the pathogenesis of cell injury. Effects of inhibitors of metabolism and membrane function on the mitochondria of Ehrlich Ascites tumor cells. Laboratory Investigation, 32:163-182. Lehninger, A.L. (1962) Water uptake and extrusion by mitochondria in relation to oxidative phosphorylation. Physiol. Rev., 424677517, Lewis, N.J., S. Akazawa, and N. Freinkel(1983)Teratogenesis from Bhydroxybutyrate during organogenesis in rat embryo organ culture and enhancement bv suhteratogenic rrlucose. Diabetes. 32:llA. Lowry, O.H., N.J. Rosebrough, A.L. k a r r , a n d R.J. Randall (1951) Protein measurement with the fohn phenol reagent. J. Biol. Chem., 193:265-275. Mahler, H.R. (1953)Role of coenzyme A in fatty acid metabolism. Fed. Proc., I2:694-702. 3-OH-BUTYRATE-INDUCEDCHANGES I N EMBRYOS McLendon, H., and J.R. Bottomy (1960) A critical analysis of the management of pregnancy in diabetic women. Am. J. Obstet. Gynecol., 80:641-649. New, D.A.T. (1978)Whole embryo culture and the study of mammalian embryos during organogenesis. Biol. Rev., 53:81-122. Pederson, M.L. (1980) Pregnancy and diabetes: A survey. Acta Endocrinol., 94:S238:13-19. Robinson, A.M., and D.H. Williamson (1980) Physiological roles of ketone bodies as substrates and signals in mammalian tissues. Physiol. Rev., 60:143-187. Romert, P., and M.E. Matthiessen (1981) Swelling of mitochondria in immersion-fixed liver tissue. Effect of various fixatives and delayed fixation. Acta Anat. (Basel),109:332-338. Sadler, T.W. (1979) Culture of early somite mouse embryos during organogenesis. J. Embryo]. Exp. Morphol., 49:17-25. 101 Sadler, T.W., and R.R. Cardell (1977) Ultrastructural alterations in neuroepithelial cells of mouse embryos exposed to cytotoxic doses of hydroxyurea. Anat. Rec., 188:103-124. Shambaugh 111, G.E., and R.A. Koehler (1983) Fetal fuels VI: Metabolism of alpha-ketoisocarproic acid in fetal rat brain. Metabolism, 32:421-427. Shambaugh 111, G.E., R.A. Koehler, and N. Freinkel(1977bj Fetal fuels 11: Contributions of selected carbon fuels to oxidative metabolism in the rat conceptus. Am. J. Physiol., 233:E457-E461. Shambaugh 111, G.E., S.C. Mrozak, and N. Freinkel(1977a) Fetal fuels I: Utilization of ketones by isolated tissues at various stages of maturation and maternal nutrition during late gestation. Metabolism, 26:623-635. Steele, C.E. (1972) Improved development of rat “egg cylinders” in vitro as a result of fusion of heart primordia. Nature, 237:150-151.